Downloaded from http://jnnp.bmj.com/ on March 8, 2015 - Published by group.bmj.com
Neurosurgery
SHORT REPORT
Deep brain stimulation for tremor resulting from acquired brain injury H A Sitsapesan,1,2 P Holland,1,2 Z Oliphant,3 N De Pennington,1,3 J-S Brittain,1,4 N Jenkinson,1,2 C Joint,1 T Z Aziz,1,2 A L Green1,2 1
Oxford Functional Neurosurgery and Experimental Neurology, Oxford University, Oxford, Oxfordshire, UK 2 Nuffield Department of Surgical Sciences, Oxford University, Oxford, Oxfordshire, UK 3 Department of Physiology, Anatomy and Genetics, Oxford University, Oxford, Oxfordshire, UK 4 Oxford University, Institute for Biomedical Engineering, Oxford, Oxfordshire, UK Correspondence to Dr Holly Sitsapesan, Nuffield Department of Surgical Sciences, Oxford University, University Offices, Level 6, West Wing, John Radcliffe Hospital, Oxford OX3 9DU, UK; [email protected] Received 5 March 2013 Revised 29 July 2013 Accepted 30 October 2013 Published Online First 4 December 2013
ABSTRACT Objectives To evaluate the efficacy of deep brain stimulation (DBS) in the treatment of tremor resulting from acquired brain injury (ABI). Methods A series of eight consecutive patients with post-ABI tremor were treated with DBS of the ventrooralis posterior (VOP)/zona incerta (ZI) region, and subsequently underwent blinded assessments using Bain's tremor severity scale. Results VOP/ZI DBS produced a mean reduction in tremor severity of 80.75% based on Bain's tremor severity scale, with significant reductions in all five component tremor subscores: rest, postural, kinetic, proximal and distal. No adverse neurological complications were reported, although one patient experienced exacerbation of pre-existing gait ataxia. Conclusion VOP/ZI stimulation is demonstrated here to be an effective and safe approach for the treatment of post-ABI tremor in the largest series published at the time of writing.
To the authors’ knowledge, this is the largest cohort study reported to date.
METHODS Patients A consecutive series of eight patients with post-ABI tremor underwent Vop/ZI DBS surgery. All patients provided informed consent for the use of anonymised data in clinical research. Table 1 displays a summary of the clinical characteristics of the patients. Patients had evidence of neural damage to midbrain or cerebellar pathways (n=6), thalamus (n=1) and peri-rolandic cortex (n=1). It is possible that patients had additional neural injury not evident on available imaging. Video assessments were part of the standard clinical assessment and postoperative follow-up. Testing was in accordance with principles of the Declaration of Helsinki and ethical approval obtained from the Oxfordshire REC C: 05/Q1605/47.
Surgery
Tremor is a debilitating comorbidity which can develop following acquired brain injury (ABI).1 2 There is well-established evidence for the efficacy of thalamic deep brain stimulation (DBS) in the treatment of tremor, particularly in Parkinson’s disease and essential tremor.3 The standard electrode target is the ventralis intermedius (Vim) thalamic nucleus. DBS is also effective for treating more complex tremor syndromes such as those associated with multiple sclerosis (MS).3–5 For complex tremor syndromes, often with proximal and distal components, DBS targeted to the ventro-oralis posterior (Vop)/zona incerta (ZI) has shown good efficacy.6 This is explained by the fact that in Vop/ZI DBS, the electrode is sited in such a way that it traverses both areas, and stimulation can be adjusted to blend the effects of Vop stimulation, which tends to target proximal tremor, and ZI stimulation, which is effective for distal tremor. Like MS tremor, post-ABI tremor is a highly variable syndrome, with heterogenous patterns of neural damage producing complex movements composed of distal and proximal components. Positive results have been reported from trials of DBS for post-ABI tremor,6–10 but mostly in the form of case reports and small series. Here we report surgical experience and follow-up of a series of eight patients diagnosed with post-ABI tremor and treated with DBS to the contralateral Vop/ZI.
Surgery was performed by one surgeon between 1999 and 2007 across two centres, Oxford, UK, and Charing Cross, UK. All patients underwent surgery for implantation of DBS electrodes (Medtronic 3387, Medtronic, Minneapolis, Minnesota, USA, each with four 1.5 mm contacts spaced 1.5 mm apart, spanning a total distance of 12 mm) targeting the Vop/ZI contralateral to the side of tremor. Details of surgical technique have previously been published by our group.5 The intended target for VOP was 12, 0, 0 (mm) relative to the MCP in each patient, while ZI was visually targeted, aiming just above the STN. The final electrode position was determined clinically by awake intraoperative testing, producing a possible discrepancy between the original target location and the final positions chosen. Stereotactic coordinates relative to the AC-PC line, electrode configuration and stimulation parameters for each patient are given in table 2. Figure 1 demonstrates the electrode trajectory (A) and the position of the two active contacts (B and C) mapped onto the Schaltenbrand–Wahren atlas for a single patient, illustrating how the electrode can traverse Vop and ZI. The typical anatomical distance between the targets varies between individuals and, if calculated linearly, also depends on the angle of approach. However, based on our subjects and the atlas localisation of the electrode contacts, the distance between ZI and Vop is between 3 and 6 mm. For example, in patient 3, where the angle of the electrode relative to the
Sitsapesan HA, et al. J Neurol Neurosurg Psychiatry 2014;85:811–815. doi:10.1136/jnnp-2013-305340
811
To cite: Sitsapesan HA, Holland P, Oliphant Z, et al. J Neurol Neurosurg Psychiatry 2014;85: 811–815.
INTRODUCTION
Downloaded from http://jnnp.bmj.com/ on March 8, 2015 - Published by group.bmj.com
Neurosurgery Table 1 Summary of clinical characteristics for patients involved in this study Case
Sex
Tremor side
Age of onset
Age at surgery
Follow-up (months)
Aetiology
Side effects
1 2
M M
L R
11 36
30 40
78 24
RTA, MRI shows cerebellar atrophy Right vertebral artery occlusion
3
F
R
23
40
46
4
F
R
40
43
12
5 6
M M
R L
11 14
30 24
78 24
7
F
R
64
68
6
Right hemisphere peri-rolandic infarction during pregnancy Right midbrain haematoma, involving right cerebellar peduncle RTA, MRI shows cerebellar atrophy RTA, diffuse axonal injury to frontal lobes +left cerebellum Ischaemic CVA, diffuse damage and large left thalamic infarct
None noted Worsening of jerk-like movements if amplitude exceeds 3.3 mA None noted
8
M
L
20
22
12
RTA, diffuse injury involving brainstem +fluid on cerebellum
None noted None noted None noted Postural tremor well suppressed with arm outstretched but more troublesome with arm bent and on intention. Right leg heaviness and dragging of leg when walking Transient pins and needles on changing settings but no other lasting symptoms
CVA, cerebrovascular accident; RTA, road traffic accident.1
AC-PC line was 49°, the centre of contact 10 was within ZI and the centre of contact 11 was within Vop, giving an estimated anatomical separation of 3 mm (see table 2 and figure 1). Electrode placement was confirmed by postoperative CT scan coregistered to the preoperative MRI. Patient 3 was additionally implanted with a second electrode in the ipsilateral globus pallidus pars interna (GPi) with the intention of suppressing dyskinesias. This lead was turned off throughout testing. Patient 3 had previously undergone a thalamotomy, which produced only transient alleviation of tremor. No adverse neurological reactions to surgery were reported. One patient suffered two infections at the implantable pulse generator (IPG) site (Medtronic Kinetra Model 7428, Medtronic, Minneapolis, Minnesota, USA) resulting in repositioning of the IPG with no further complications. One patient suffered an IPG failure resulting in rapid tremor recurrence, which was abolished upon IPG replacement. During follow-up,
patients 3, 5 and 6 had replacement of IPG due to end of battery of life.
Clinical rating scale The severity of tremor was measured using Bain’s standardised clinical rating scale for tremor.11 Anonymised video clips of clinical assessments were digitised, randomised and rated by two consultant members of the clinical care team blinded to the patient’s operative and stimulation status. Postoperative videos from the longest follow-up period available were selected. Ratings addressed severity of tremor in the affected arm in five categories: rest, postural, kinetic, proximal and distal. Ratings were scored out of 10 for each category with 0 representing the absence of tremor and 10 being the most severe. Ratings were based on the amplitude and intermittency of tremor. Clinical definitions of tremor components have previously been established and described
Table 2 Active electrode contacts, stimulation parameters and stereotactic coordinates for each patient where data are available
Stimulation parameters
Angle between e lectrode and AC-PC line in sagittal plane
Stereotactic coordinates of active contacts relative to AC-PC line
0 in ZI
2.0 V, 90 μs, 130 Hz
64°
1 in ZI, 2 and 3 in VOP
2.5 V, 90 μs, 130 Hz
68°
7.3 posterior, 12.2 left, 2.5 inferior 2.0 posterior, 14.5 left, 6.3 superior 0.8 anterior, 10.4 left, 4.6 superior 0.4 posterior, 9.1 left, 2.5 superior 3.2 posterior, 12.0 right, 1.6 inferior 2.8 posterior, 12.6 right, 1.9 superior 5.1 posterior, 13.1 right, 2.5 inferior 3.7 posterior, 14.7 right, 2.9 superior – – – 3.1 posterior, 10.8 right, 3.2 inferior 0.7 posterior, 13.9 right, 2.3 superior
Electrode configuration
Electrode contacts in the target areas
Patient
Electrode side
1
Left
2
Right
Case (+), 3 (−) (monopolar) 3 (+), 2 (−)
3
Right
9 (−), 10 (+)
10 in ZI, 11 in VOP
3.0 V, 210 μs, 100 Hz
49°
4
Right
4 (−), 6 (+)
1.5 V, 90 μs, 130 Hz
80°
5 6 7 8
Right Left Right Left
5 (−), 7 (+) 3 (+), 2 (−) 0 (−), 2 (+) 1 (+), 3 (−)
4 in ZI, 6 and 7 adjacent to VOP – – – 1 in ZI, 2 and 3 in VOP
2.3 V, 3.6 V, 2.6 V, 3.3 V,
210 μs, 180 Hz 360 μs, 130 Hz 300 μs, 130 Hz 156 μs, 130 Hz
– – – 60°
A dash indicates unavailable data due to missing scans. Numbering of contacts is based on two different systems; older devices have contacts numbered 0–3 and 4–7 while newer devices have contacts numbered 0–3 and 8–11. Column 4 shows which contacts fall within the target area based on the Schaltenbrand–Wahren atlas. VOP, ventro-oralis posterior; ZI, zona incerta.
812
Sitsapesan HA, et al. J Neurol Neurosurg Psychiatry 2014;85:811–815. doi:10.1136/jnnp-2013-305340
Downloaded from http://jnnp.bmj.com/ on March 8, 2015 - Published by group.bmj.com
Neurosurgery Figure 1 Electrode trajectory (A) and location of active contacts (white circle indicated by white arrow) for a single patient ( patient 3) displayed on Schaltenbrand–Wahren atlas. (B) Contact 11, located in ventro-oralis posterior and (C) contact 10 in zona incerta.
in detail.12 The mean score from the two observers was used for statistical testing.
RESULTS All patients were assessed and videoed ON stimulation (n=8) at a mean length of follow-up of 26 months. ON stimulation tremor scores were compared with either OFF stimulation scoring of videos made at the time of follow-up assessment or scores from preoperative videos. Six patients had preoperative video assessments for comparison and five patients had follow-up OFF stimulation video assessments taken after a sufficient wash-out period. Preoperative videos were used in preference over follow-up videos in the six patients where these were available. In the patient with a GPi lead, this was turned OFF throughout all assessments. For the six patients with preoperative and ON stimulation assessments, all showed reductions in
tremor severity across the five components of the Bain scale (mean reduction 69.13%, SD 38.69%). Similar results were obtained for the two patients in which ON and OFF stimulation conditions were instead compared ( patient 7: mean reduction 77.78% and patient 5: mean reduction 87%). Taken together, there was an overall reduction of tremor severity by 80.75% following stimulation. Figure 2 displays the mean component scores across conditions ( preoperative, ON stimulation, OFF stimulation) and table 3 gives individual scores for patients 1–7 (scores for patient 8 are unavailable at the time of writing). The scores from the two observers showed a highly significant correlation (n=95, p
Neurosurgery
SHORT REPORT
Deep brain stimulation for tremor resulting from acquired brain injury H A Sitsapesan,1,2 P Holland,1,2 Z Oliphant,3 N De Pennington,1,3 J-S Brittain,1,4 N Jenkinson,1,2 C Joint,1 T Z Aziz,1,2 A L Green1,2 1
Oxford Functional Neurosurgery and Experimental Neurology, Oxford University, Oxford, Oxfordshire, UK 2 Nuffield Department of Surgical Sciences, Oxford University, Oxford, Oxfordshire, UK 3 Department of Physiology, Anatomy and Genetics, Oxford University, Oxford, Oxfordshire, UK 4 Oxford University, Institute for Biomedical Engineering, Oxford, Oxfordshire, UK Correspondence to Dr Holly Sitsapesan, Nuffield Department of Surgical Sciences, Oxford University, University Offices, Level 6, West Wing, John Radcliffe Hospital, Oxford OX3 9DU, UK; [email protected] Received 5 March 2013 Revised 29 July 2013 Accepted 30 October 2013 Published Online First 4 December 2013
ABSTRACT Objectives To evaluate the efficacy of deep brain stimulation (DBS) in the treatment of tremor resulting from acquired brain injury (ABI). Methods A series of eight consecutive patients with post-ABI tremor were treated with DBS of the ventrooralis posterior (VOP)/zona incerta (ZI) region, and subsequently underwent blinded assessments using Bain's tremor severity scale. Results VOP/ZI DBS produced a mean reduction in tremor severity of 80.75% based on Bain's tremor severity scale, with significant reductions in all five component tremor subscores: rest, postural, kinetic, proximal and distal. No adverse neurological complications were reported, although one patient experienced exacerbation of pre-existing gait ataxia. Conclusion VOP/ZI stimulation is demonstrated here to be an effective and safe approach for the treatment of post-ABI tremor in the largest series published at the time of writing.
To the authors’ knowledge, this is the largest cohort study reported to date.
METHODS Patients A consecutive series of eight patients with post-ABI tremor underwent Vop/ZI DBS surgery. All patients provided informed consent for the use of anonymised data in clinical research. Table 1 displays a summary of the clinical characteristics of the patients. Patients had evidence of neural damage to midbrain or cerebellar pathways (n=6), thalamus (n=1) and peri-rolandic cortex (n=1). It is possible that patients had additional neural injury not evident on available imaging. Video assessments were part of the standard clinical assessment and postoperative follow-up. Testing was in accordance with principles of the Declaration of Helsinki and ethical approval obtained from the Oxfordshire REC C: 05/Q1605/47.
Surgery
Tremor is a debilitating comorbidity which can develop following acquired brain injury (ABI).1 2 There is well-established evidence for the efficacy of thalamic deep brain stimulation (DBS) in the treatment of tremor, particularly in Parkinson’s disease and essential tremor.3 The standard electrode target is the ventralis intermedius (Vim) thalamic nucleus. DBS is also effective for treating more complex tremor syndromes such as those associated with multiple sclerosis (MS).3–5 For complex tremor syndromes, often with proximal and distal components, DBS targeted to the ventro-oralis posterior (Vop)/zona incerta (ZI) has shown good efficacy.6 This is explained by the fact that in Vop/ZI DBS, the electrode is sited in such a way that it traverses both areas, and stimulation can be adjusted to blend the effects of Vop stimulation, which tends to target proximal tremor, and ZI stimulation, which is effective for distal tremor. Like MS tremor, post-ABI tremor is a highly variable syndrome, with heterogenous patterns of neural damage producing complex movements composed of distal and proximal components. Positive results have been reported from trials of DBS for post-ABI tremor,6–10 but mostly in the form of case reports and small series. Here we report surgical experience and follow-up of a series of eight patients diagnosed with post-ABI tremor and treated with DBS to the contralateral Vop/ZI.
Surgery was performed by one surgeon between 1999 and 2007 across two centres, Oxford, UK, and Charing Cross, UK. All patients underwent surgery for implantation of DBS electrodes (Medtronic 3387, Medtronic, Minneapolis, Minnesota, USA, each with four 1.5 mm contacts spaced 1.5 mm apart, spanning a total distance of 12 mm) targeting the Vop/ZI contralateral to the side of tremor. Details of surgical technique have previously been published by our group.5 The intended target for VOP was 12, 0, 0 (mm) relative to the MCP in each patient, while ZI was visually targeted, aiming just above the STN. The final electrode position was determined clinically by awake intraoperative testing, producing a possible discrepancy between the original target location and the final positions chosen. Stereotactic coordinates relative to the AC-PC line, electrode configuration and stimulation parameters for each patient are given in table 2. Figure 1 demonstrates the electrode trajectory (A) and the position of the two active contacts (B and C) mapped onto the Schaltenbrand–Wahren atlas for a single patient, illustrating how the electrode can traverse Vop and ZI. The typical anatomical distance between the targets varies between individuals and, if calculated linearly, also depends on the angle of approach. However, based on our subjects and the atlas localisation of the electrode contacts, the distance between ZI and Vop is between 3 and 6 mm. For example, in patient 3, where the angle of the electrode relative to the
Sitsapesan HA, et al. J Neurol Neurosurg Psychiatry 2014;85:811–815. doi:10.1136/jnnp-2013-305340
811
To cite: Sitsapesan HA, Holland P, Oliphant Z, et al. J Neurol Neurosurg Psychiatry 2014;85: 811–815.
INTRODUCTION
Downloaded from http://jnnp.bmj.com/ on March 8, 2015 - Published by group.bmj.com
Neurosurgery Table 1 Summary of clinical characteristics for patients involved in this study Case
Sex
Tremor side
Age of onset
Age at surgery
Follow-up (months)
Aetiology
Side effects
1 2
M M
L R
11 36
30 40
78 24
RTA, MRI shows cerebellar atrophy Right vertebral artery occlusion
3
F
R
23
40
46
4
F
R
40
43
12
5 6
M M
R L
11 14
30 24
78 24
7
F
R
64
68
6
Right hemisphere peri-rolandic infarction during pregnancy Right midbrain haematoma, involving right cerebellar peduncle RTA, MRI shows cerebellar atrophy RTA, diffuse axonal injury to frontal lobes +left cerebellum Ischaemic CVA, diffuse damage and large left thalamic infarct
None noted Worsening of jerk-like movements if amplitude exceeds 3.3 mA None noted
8
M
L
20
22
12
RTA, diffuse injury involving brainstem +fluid on cerebellum
None noted None noted None noted Postural tremor well suppressed with arm outstretched but more troublesome with arm bent and on intention. Right leg heaviness and dragging of leg when walking Transient pins and needles on changing settings but no other lasting symptoms
CVA, cerebrovascular accident; RTA, road traffic accident.1
AC-PC line was 49°, the centre of contact 10 was within ZI and the centre of contact 11 was within Vop, giving an estimated anatomical separation of 3 mm (see table 2 and figure 1). Electrode placement was confirmed by postoperative CT scan coregistered to the preoperative MRI. Patient 3 was additionally implanted with a second electrode in the ipsilateral globus pallidus pars interna (GPi) with the intention of suppressing dyskinesias. This lead was turned off throughout testing. Patient 3 had previously undergone a thalamotomy, which produced only transient alleviation of tremor. No adverse neurological reactions to surgery were reported. One patient suffered two infections at the implantable pulse generator (IPG) site (Medtronic Kinetra Model 7428, Medtronic, Minneapolis, Minnesota, USA) resulting in repositioning of the IPG with no further complications. One patient suffered an IPG failure resulting in rapid tremor recurrence, which was abolished upon IPG replacement. During follow-up,
patients 3, 5 and 6 had replacement of IPG due to end of battery of life.
Clinical rating scale The severity of tremor was measured using Bain’s standardised clinical rating scale for tremor.11 Anonymised video clips of clinical assessments were digitised, randomised and rated by two consultant members of the clinical care team blinded to the patient’s operative and stimulation status. Postoperative videos from the longest follow-up period available were selected. Ratings addressed severity of tremor in the affected arm in five categories: rest, postural, kinetic, proximal and distal. Ratings were scored out of 10 for each category with 0 representing the absence of tremor and 10 being the most severe. Ratings were based on the amplitude and intermittency of tremor. Clinical definitions of tremor components have previously been established and described
Table 2 Active electrode contacts, stimulation parameters and stereotactic coordinates for each patient where data are available
Stimulation parameters
Angle between e lectrode and AC-PC line in sagittal plane
Stereotactic coordinates of active contacts relative to AC-PC line
0 in ZI
2.0 V, 90 μs, 130 Hz
64°
1 in ZI, 2 and 3 in VOP
2.5 V, 90 μs, 130 Hz
68°
7.3 posterior, 12.2 left, 2.5 inferior 2.0 posterior, 14.5 left, 6.3 superior 0.8 anterior, 10.4 left, 4.6 superior 0.4 posterior, 9.1 left, 2.5 superior 3.2 posterior, 12.0 right, 1.6 inferior 2.8 posterior, 12.6 right, 1.9 superior 5.1 posterior, 13.1 right, 2.5 inferior 3.7 posterior, 14.7 right, 2.9 superior – – – 3.1 posterior, 10.8 right, 3.2 inferior 0.7 posterior, 13.9 right, 2.3 superior
Electrode configuration
Electrode contacts in the target areas
Patient
Electrode side
1
Left
2
Right
Case (+), 3 (−) (monopolar) 3 (+), 2 (−)
3
Right
9 (−), 10 (+)
10 in ZI, 11 in VOP
3.0 V, 210 μs, 100 Hz
49°
4
Right
4 (−), 6 (+)
1.5 V, 90 μs, 130 Hz
80°
5 6 7 8
Right Left Right Left
5 (−), 7 (+) 3 (+), 2 (−) 0 (−), 2 (+) 1 (+), 3 (−)
4 in ZI, 6 and 7 adjacent to VOP – – – 1 in ZI, 2 and 3 in VOP
2.3 V, 3.6 V, 2.6 V, 3.3 V,
210 μs, 180 Hz 360 μs, 130 Hz 300 μs, 130 Hz 156 μs, 130 Hz
– – – 60°
A dash indicates unavailable data due to missing scans. Numbering of contacts is based on two different systems; older devices have contacts numbered 0–3 and 4–7 while newer devices have contacts numbered 0–3 and 8–11. Column 4 shows which contacts fall within the target area based on the Schaltenbrand–Wahren atlas. VOP, ventro-oralis posterior; ZI, zona incerta.
812
Sitsapesan HA, et al. J Neurol Neurosurg Psychiatry 2014;85:811–815. doi:10.1136/jnnp-2013-305340
Downloaded from http://jnnp.bmj.com/ on March 8, 2015 - Published by group.bmj.com
Neurosurgery Figure 1 Electrode trajectory (A) and location of active contacts (white circle indicated by white arrow) for a single patient ( patient 3) displayed on Schaltenbrand–Wahren atlas. (B) Contact 11, located in ventro-oralis posterior and (C) contact 10 in zona incerta.
in detail.12 The mean score from the two observers was used for statistical testing.
RESULTS All patients were assessed and videoed ON stimulation (n=8) at a mean length of follow-up of 26 months. ON stimulation tremor scores were compared with either OFF stimulation scoring of videos made at the time of follow-up assessment or scores from preoperative videos. Six patients had preoperative video assessments for comparison and five patients had follow-up OFF stimulation video assessments taken after a sufficient wash-out period. Preoperative videos were used in preference over follow-up videos in the six patients where these were available. In the patient with a GPi lead, this was turned OFF throughout all assessments. For the six patients with preoperative and ON stimulation assessments, all showed reductions in
tremor severity across the five components of the Bain scale (mean reduction 69.13%, SD 38.69%). Similar results were obtained for the two patients in which ON and OFF stimulation conditions were instead compared ( patient 7: mean reduction 77.78% and patient 5: mean reduction 87%). Taken together, there was an overall reduction of tremor severity by 80.75% following stimulation. Figure 2 displays the mean component scores across conditions ( preoperative, ON stimulation, OFF stimulation) and table 3 gives individual scores for patients 1–7 (scores for patient 8 are unavailable at the time of writing). The scores from the two observers showed a highly significant correlation (n=95, p
